Electrical Power and Energy Systems 99 (2018) 175–182

Contents lists available at ScienceDirect

Electrical Power and Energy Systems

journal homepage: www.elsevier.com/locate/ijepes

A Markovian model for power transformer maintenance T

⁎
Zhenglin Liang , Ajith Parlikad
Asset management group, Institute for Manufacturing at the University of Cambridge, UK

A R T I C L E I N F O A B S T R A C T

Keywords: The condition of the insulation paper is one of the key determinants of the lifetime of a power transformer. The
Value of information winding insulation paper may deteriorate aggressively and result in the unexpected failure of power transfor-
Condition-based maintenance mers, especially under the presence of high moisture, oxygen, and metal contaminants. Such types of scenarios
Reliability engineering and power transformer can be prevented if the deterioration is detected on time. Various types of condition monitoring techniques have
been developed to detect transformer condition such as dissolved gas analysis (DGA) and frequency response
analysis (FRA). They are non-intrusive and provide early warning of accelerated deterioration both chemically
and mechanically. However, the accuracy of those techniques is imperfect, which means periodic inspection is
still indispensable. In this paper, we discuss the value of continuous condition monitoring for power transformers
and present a way to estimate this value. Towards this, a continuous-time Markov decision model is presented to
optimize periodic inspections, so that the cost is minimized and the availability is maximized. We then analyze
the performance based on the information from both discrete inspection and continuous condition monitoring
using DGA and FRA. The result shows the dissolved gas analysis can improve the availability and operation cost,
while frequency response analysis can only improve the availability of power transformers.

1. Introduction condition [6]. Abiri-Jahromi et al. have developed a two-stages main-

Power transformers are critical assets in a power transmission net-
work. A failure of a power transformer also may cause cascading failure
and catastrophic blackout in the power grid. The necessity of increasing
reliability and availability of power transformers can be analyzed di-
rectly from a financial point of view. Between 1997 and 2001, the total
losses caused by power transformer failure in the US were over
286 million [1]. Moreover, the aging population of power transformers
has increased since 1975 [2]. These imply that it is expected to have an
increase in power transformer failure, and the resulting load curtail-
ment if the maintenance strategy remains the same.
In literature, various types of maintenance models have been de-
veloped to address the problem of power transformer maintenance.
Aldhubaib and Salama have developed a reliability centered main-
tenance and replacement approach to optimize maintenance and re-
placement to increase the lifetime of power transformers and reduce
annual cost [3]. Dhople et al. proposed a set-theoretic method for
capturing the uncertainty in Markov reliability and reward model to
maximize the availability of power transformers [4]. Abu-Elanien et al.
developed a decision support system to determine the life expectancy of
transformers from techno-economic perspective [5]. Lima et al. de-
signed a two-level framework of fault diagnosis and decision making for
power transformers with considering the loss for life caused by overload
tenance management model that contains both mid-term and short-
term maintenance to maximize the serviceability of power transformers
[7]. Koksal and Ozdemir have improved the power transformer main-
tenance using a Markovian model [8].
As to the condition state of a power transformer is considered to be
discrete, most of the developed models are based on the Markovian
deterioration model. However, the deterioration of power transformers
is oversimplified and modeled by Markov chain with a single dete-
riorating path. Such an approach is inaccurate because it overlooks the
complexity of the deterioration of power transformers, such as the ac-
celeration deterioration of insulation paper caused by high moisture.
Because of this, the effective of condition-based maintenance that solely
relies on the periodic inspection is over-estimated. Therefore, to re-es-
timate the value of continuous monitoring, it is essential to improve the
deterioration model of power transformers. In practice, the accuracy of
the condition monitoring is imperfect and may be interfered by op-
eration signals and external signals. Therefore, even for the power
transformers that have already installed the condition monitoring de-
vices, periodic inspection can still provide additional value to triangu-
late the estimated condition information by condition monitoring. The
objective of the paper is twofold: optimize the condition-based main-
tenance for power transformers; explore the value of online monitoring
from the perspective of the lifecycle of power transformers.

⁎
Corresponding author at: Institute for Manufacturing, 17 Charles Babbage Rd, Cambridge CB3 0FS, UK.
E-mail address: zl284@cam.ac.uk (Z. Liang).

https://doi.org/10.1016/j.ijepes.2017.12.024
Received 17 September 2017; Received in revised form 29 November 2017; Accepted 25 December 2017
Available online 20 February 2018
0142-0615/ © 2018 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/BY/4.0/).
Z. Liang, A. Parlikad
Nomenclature
πij steady state probability of state (i,j )
λa infant mortality rate of power transformers
the deterioration rate of winding insulation paper from
state (i,0) at normal deterioration process
λdi,1 the deterioration rate of winding insulation paper from
state (i,1) in the accelerated chemical deterioration
λdi,2 the deterioration rate of winding insulation paper from
state (i,2) in the accelerated mechanical deterioration
λ fi,1 the transition rate from state (i,0) in the normal dete-
rioration to state (i,1) in the accelerated chemical dete-
rioration
λ fi,2 the transition rate from state (i,0) in the normal dete-
rioration to state (i,2) in the accelerated mechanical de-
terioration
λF the sudden failure rate of power transformer in the normal
deterioration process
λFd the sudden failure rate of power transformer in the
To achieve the objectives, in the second section, we use cause-effect
analysis on different subsystems of the power transformer to identify
the potential risk of acceleration deterioration of insulation paper
caused by the malfunction of different subsystems. In section three, we
develop a continuous-time Markov chain model to optimize the main-
tenance of power transformers based on the information from inspec-
tion and continuous monitoring. Section four analyzes the value of
different types of online monitoring numerically. Section five sum-
marizes the concluding remarks of the paper.
2. Deterioration of power transformers
To systematically analyze the deterioration of power transformers to
identify the information that can be used to improve the modeling of
the power transformer. According to functionality and structure, power
transformers can be classified into seven subsystems: winding, magnetic
core, insulation oil, bushing, tap changer, tank and cooling equipment.
In practice, the condition of winding insulation paper is usually re-
garded as the index for power transformer condition [9] and [10]. The
deterioration of the winding insulation paper may accelerate under the
presence of high moisture, oxygen, and metal contaminants.
In this section, we aim to identify the potential risks of the ac-
celerated deterioration of winding insulation paper caused by the
malfunctions of other subsystems using cause-effect analysis. In gen-
eral, the deterioration of winding insulation paper may accelerate in
two ways: accelerated chemical aging and accelerated mechanical
Fig. 1. Malfunctions of subsystems that can cause an accelerated deterioration of winding insulation.
176
Electrical Power and Energy Systems 99 (2018) 175–182
accelerated deterioration process
λ ol the rate of successfully detected malfunctions by online
monitoring device
1/ μin the duration of periodic inspection
1/ μc the duration of minor maintenance
1/ μM the duration of major maintenance
1/ μF the duration of corrective maintenance
1/ μR the duration of replacement
Cu the downtime penalty cost due to unexpected failure
Cp The downtime penalty cost due to maintenance and in-
spection
Cin′ the cost of periodic inspection
CC′ the cost of minor maintenance
CM′ The cost of major maintenance
CF′ the cost of corrective maintenance
CR′ the cost of replacement
Col the annual cost of online monitoring device
aging. The accelerated chemical deterioration is a combination of three
interactive processes: pyrolysis, hydrolysis, and oxidation [11]. Hy-
drolysis is the dominant process in the accelerated chemical degrada-
tion. The rate of hydrolysis is dependent on the content of moisture and
catalyzes by the acidity [12]. The increase of acidity is caused by the
sludge formation as the result of oxidation. The sludge will also increase
the temperature and accelerate the pyrolysis. The accelerated chemical
deterioration starts with the occurrence of contamination and moderate
partial discharge. During the deterioration process, dissolved gas will
be generated. Eventually, the partial discharge will result in treeing,
tracking or even breakdown the winding insulation. The accelerated
mechanical deterioration is usually initialized by the loss of clamping
force or distortion of winding geometry, which is mainly caused by
abrasion under electric-magnetic forces [13]. Under accelerated me-
chanical deterioration, the partial discharge will proceed to creeping
and result in the breakdown of winding insulation.
The aging of winding insulation is related to the moisture, acidity,
oxygen, containment level, and clamping forces. Empirically, abnorm-
ality in these factors is usually caused by malfunctions of other sub-
systems. For example, inelastic gasket on bushing can increase the risk
of excessive moisture, oxygen, and containment level and in turn ac-
celerate the rate of winding insulation aging and reduce the life of the
power transformer. Malfunctions such as inelastic gasket can be re-
paired with a minor cost if it is detected on time. However, the resulting
deterioration of the insulation paper is irreversible and will sig-
nificantly reduce the service lifetime of the power transformer.
Z. Liang, A. Parlikad
Therefore, it is valuable to identify the malfunctions on different sub-
systems that may result in the accelerated deterioration of the insula-
tion paper. We have identified the malfunctions from the extant lit-
erature using cause-effect analysis. The result is illustrated in Fig. 1.
In practice, the malfunction rates of the subsystems of power
transformer are recorded as shown in [14]. As a result of the findings in
this section, we can identify the transition rate between normal dete-
rioration and accelerated deterioration for the winding insulation paper
and refine the deterioration model of power transformers. In the next
section, we will apply this knowledge to formulate the deterioration
and maintenance model of power transformers in both distribution
networks and transmission networks.
3. Deterioration and maintenance models for power transformers
We use continuous-time Markov chain (CTMC) to model the dete-
rioration and maintenance of power transformers. Practically speaking,
the maintenance strategy used for power transformers in distribution
networks and those in transmission networks are different. We first
model the deterioration and maintenance strategy for power transfor-
mers in distribution networks. The state transition diagram is illustrated
in Fig. 2.
In Fig. 2, the states are represented by two indices i and j . i is the
index representing the condition of the winding insulation paper and j
is the index that represents the type of deterioration. We classify the
condition of winding insulation i into 5 states (healthy, aged, defective,
faulty and failure). Normal deterioration process is indicated with
j = 0 , while the accelerated chemical aging is presented as j = 1 and
accelerated mechanical aging is represented as j = 2 . We denote λ fi,j as
the rate of transition from condition state (i,0) to (i,j ) . In an accelerated
deterioration process, power transformer reaches the deterioration
failure state (4,0) faster than normal process. Apart from deterioration
failure, we also consider sudden failure with probability λF , which is
mainly caused by some exogenous events. In practice, if one of the
power transformer subsystems has malfunctioned, the overall vulner-
ability of transformer will rise. We represent the probability of sudden
failure during accelerated deterioration as λFd (λFd ⩾ λF ). The infant
mortality rate λa is also considered in the first condition state. The
combination of all failure rates is a bathtub curve with diverse wear-out
tails as shown in Fig. 3.
In Fig. 3, the wear-out tail is not only dependent on the type of the
accelerated deterioration process it undergoes, but also on the timing of
when the malfunctions result in accelerated deterioration.
In some scenarios, the main objective of maintenance of power
transformers in the distribution network is to retain its basic service-
ability. Corrective maintenance is implemented after the sudden failure
of a power transformer. It can effectively restore the transformer to the
normal operating condition just before the failure. However, not all
failures are repairable. After the breakdown of winding insulation, re-
placement of the power transformer is necessary to preserve the ser-
viceability of the power transformer. We denote the duration of cor-
rective maintenance as 1/ μF with a cost CF′ , and replacement time as
1/ μR with a cost CR′ . Meanwhile, failure also results in unplanned
downtime penalty Cu due to the risk of load curtailment. The avail-
ability of the power transformer A is shown in Eq. (1):
3
A = 1−μF ∑ πi,3 + μR π4,0
i=0
where πi,j indicates the steady state probability of the state (i,j ) . The
long-term operational cost C is shown in Eq. (2).
3 3
⎛ ⎞
C = Cu ⎜∑ πi,3 + π4,0⎟ + CF′ μF ∑ πi,3 + CR′ μR π4,0
⎝ i = 0 ⎠ i=0
In Eq. (2), the operation cost is a sum of three items. The first one is the
downtime penalty cost for load curtailment. The second is the cost of
Electrical Power and Energy Systems 99 (2018) 175–182
corrective maintenance and the third is the cost of replacement. To
derive the analytic solution for Eqs. (1) and (2), we need to express the
steady state probabilities analytically. Firstly, this requires expressing
all states with respect to a reference state for which we choose π0,0 . The
expressions of all the states in term of π0,0 is shown in Appendix A. By
recalling that the sum of all the steady state probabilities is equal to 1,
we can calculate π0,0 and in turn calculate the rest of steady state
probabilities.
For power transformers in transmission networks, failure can be
catastrophic and is associated with huge penalty costs. It is important to
prevent failures and maximize the availability of power transformers.
Here, condition-based maintenance is applied in addition to corrective
maintenance and replacement. Condition-based maintenance is a type
of maintenance strategy that recommends maintenance activities based
on the information acquired through periodic inspection or online
monitoring. In practice, both periodic inspection and online monitoring
have their limitations. In periodic inspection, the deterioration may
proceed to an unacceptable state or even failure state between the two
successive inspections because of the stochastic nature of aging. The
limitation of online monitoring is that not all types of malfunctions can
be detected by online monitoring. Therefore, it is practical to imple-
ment both types of monitoring on power transformers in transmission
networks.
The state transition diagram for condition-based maintenance of
power transformers in transmission networks is shown in Fig. 4. The
rate of periodic inspection is denoted as λin . If the inspection indicates
that the power transformer is in an acceptable condition (i < 3 at
normal deterioration), it will lead to no further action as presented by
state (i,4) . If any type of accelerated deterioration of the winding in-
sulation is detected, a minor maintenance will be applied as represented
by state (i,5) . The malfunctioned subsystems will be repaired and de-
terioration will recover back to normal thereby prolonging the service
lifetime of the transformer. However, because the deterioration of
winding insulation paper is irreversible, minor maintenance will not
restore the condition of the winding insulation. We denote the state of
minor maintenance as state (i,6) where i < 3. The duration of the minor
maintenance is 1/μ c with a cost Cc′. If i = 3, a major maintenance (3,6)
will be implemented to completely overhaul the power transformer.
The major maintenance includes the replacement of the deteriorated
winding insulation paper and can restore the condition of transformer
back to (0,0) . The duration of major maintenance is 1/ μM with a cost CM ′.
State (i,7) indicates the minor preventive maintenance that is im-
mediately scheduled after an early warning signal of a malfunction is
detected by online monitoring. The rate of such type of preventive
maintenance is λ ol . Because of preventive maintenance, the rates of
malfunction λ fi,1 or λ fi,2 are reduced. When the rate λ ol = 0 , the state
diagram indicates the scenario when maintenance strategy is based on
the periodic inspection only, without the support of online monitoring.
(1)
(2)
Fig. 2. State transition diagram for deterioration and maintenance of power transformers
in distribution networks.
177
Z. Liang, A. Parlikad Electrical Power and Energy Systems 99 (2018) 175–182

3 2 3 7
⎛ ⎞ ⎛ ⎞
C = Cu ⎜∑ πi,3 + π4,0⎟ + Cp ∑ πi,4 + ∑∑ πi,j
⎜ ⎟
⎝ i=0 ⎠ ⎝ i=0 i=0 j=5
⎠
2 3 2 3
⎛ ⎞ ⎛ ⎞
+ Cin′ μin ⎜∑ πi,4 + ∑ πi,5⎟ + Cc′ μc ⎜∑ πi,6 + ∑ ′ μM π3,6
πi,7⎟ + CM
⎝ i=0 i=0 ⎠ ⎝ i=0 i=0 ⎠
3
+ CF′ μF ∑ πi,3 + CR′ μR π4,0 + Col
i=0 (4)

Fig. 3. A bathtub curve for the combination of different types of failure rates.
When online monitoring is implemented, early warning signals of
some malfunctions can be detected on time. As a result, the malfunction
can be repaired by a minor maintenance preventively before causing
any non-negligible damage to the winding insulation. (i,7) represents
the states for such type of minor preventive maintenances with a rate
λ ol . In this case, the rates of malfunctions are reduced due to the pre-
ventive maintenance that are enabled by online monitoring. Thus, λ ol is
equal to the reduced malfunction rate. We denote the cost of online
monitoring as Col . We assume that all the maintenance and inspection
activities will require the transformer to be stopped. The penalty cost
per unit time for such stoppages is denoted as Cp . Because maintenance
and inspection are preventive and prescheduled activities, Cp is sig-
nificantly smaller than Cu , which is caused by unexpected failures. With
the similar method used for the distribution network, we can express
the availability and operation cost respectively as shown in Eqs. (3) and
(4).
3 2
A= ∑∑ πi,j
i=0 j=0
In Eq. (4), the operation cost is the sum of the downtime penalty cost
for unexpected failure, downtime penalty cost for preventive activities,
inspection cost, minor maintenance cost, major maintenance cost,
correction maintenance cost, replacement cost and the cost of online
monitoring.This condition-based maintenance model opens up a suc-
cessful way to find the optimal mean time between inspections to
maximize the availability and minimize the operation cost with or
without online monitoring. In Section 4, a numerical case study is
provided to demonstrate the approach and to assess the value of dif-
ferent types of online monitoring.
4. Value of monitoring
In general, online monitoring is worthwhile only for power trans-
formers in transmission networks, and we restrict our focus to those.
The illustrative example is based on the parameter settings of 220 kV
oil-immersed power transformers. We first aim to find the optimal mean
time between inspections to maximize availability or minimize opera-
tion cost without applying online monitoring. Then, we optimize the
mean time between inspections with different types of online mon-
itoring. Using that, we assess the value of different types of online
monitoring.
To ease the understanding, we assume the deterioration rates of
winding insulation paper are independent of condition states. Degree of
(3) polymerization (DP) is widely used as the measurement of the condition
of insulation paper. As shown in [15], 1/DP can be used as the condi-
tion boundaries for the aging insulation paper. Based on [16], the
condition of insulation paper reaches the failure state when DP is below

Fig. 4. State transition diagram for condition-based

maintenance of power transformers in transmission
networks.

178
Z. Liang, A. Parlikad Electrical Power and Energy Systems 99 (2018) 175–182

200. In our case, the condition states of Markov chain are classified From Fig. 8, we can see that the maximum availability is 99.48%
into five states using the DP as condition boundaries (1000–500, with DGA when the mean time between inspections is about 2 years.
500–333, 333–250, 250–200 and below 200). We reduce the Compared to the optimized availability without DGA, the improvement
number of parameters by using λn = λn0,0 = λn0,1 = λn0,2 = λn0,3, by applying DGA is 0.6%. When the mean time between inspections is
λd1 = λd0,1 = λ d1,1 = λ d2,1 = λd3,1, λ d2 = λ d0,2 = λ d1,2 = λ d2,2 = λ d3,2 , smaller than 0.4 years, little improvement can be made by using DGA.
λ f 1 = λ f 0,1 = λ f 1,1 = λ f 2,1 = λ f 3,1 and λ f 2 = λ f 0,2 = λ f 1,2 = λ f 2,2 = λ f 3,2 . The improvement in availability increases with increase in mean time
First, we consider the scenario without the online monitoring. The between inspections.
scenario is represented in the model as λ ol = 0 and Col = 0 . The para- Fig. 9 shows the operation cost under different mean time between
meters λ f 1 and λ f 2 are evaluated by experts according to the cause-effect inspections with DGA. The minimum operation cost is £85,895/year
analysis in Section 2. The overall numerical setting for all the para- with a mean time between inspection of 0.685 years. Compared to the
meters is shown in Table 1. In the table, the values of parameters for optimized operation cost without DGA, the savings is about £3855/
deterioration and maintenance duration are estimated based on the year. It is worthwhile to notice that when the mean time between in-
secondary data with the verification of experts. In practice, the values spection is smaller than 0.2 years, the operation cost without DGA is
of failure cost and maintenance cost may vary case by case. smaller due to the cost of online monitoring. The average rate of major
Based on the parameter settings, we can calculate the availability maintenance is reduced to 0.0504/year.
and cost of the power transformer using Eqs. (3) and (4). We plot the We will then investigate the performance of FRA. The accuracy of
availability and operation cost against the mean time between inspec- FRA to detect the mechanical related malfunctions is shown to be 90%
tions in Figs. 5 and 6 respectively. in [23]. In this scenario, the λ ol is:
Because both availability and operation cost are unimodal, we can
λ ol = 90% × λ f 2 = 0.0009/ year
find the optimal mean time between inspection to maximize the
availability using Eq. (5). and the λ f 2 is reduced to 0.0001/year. We assume Col remain the same.
dA The resulting availability and operation cost are illustrated in the
=0
dλin (5) Figs. 10 and 11.
In Fig. 10, the maximized availability is 99.43% with FRA when the
Likewise, we can calculate the optimal mean time between inspection mean time between inspection is about 1.86 years. Compared to the
to minimize the operation cost using Eq. (6). optimized availability without FRA, the improvement by applying FRA
dC is 0.01%.
=0
dλin (6) Fig. 11 shows the minimized operation cost for the power trans-
former with FRA is £91,840/year. It is even larger than the scenario
It can be found from Fig. 5 that the optimal mean time between in- without FRA. This is because the performance improvement cannot
spections to maximize the availability is about 1.7 years and the max- surpass the annual cost of online monitoring. The average rate of major
imum availability is 99.42%. From Fig. 6, the operation cost is mini- maintenance is reduced to 0.0517/year.
mized at £89,750/year, when mean time between inspections is around Table 3 compares the performance between DGA and FRA
0.33 years. The average rate of major maintenance is 0.0519/year. As shown in Table 3, DGA can improve the availability, reduce the
′ and CR′ would be smaller than
In practice, the cost setting of Cu,Cp,CM operation cost and reduce the average rate of major maintenance. But,
the default setting because protection systems, control algorithms and FRA only provides a relatively small improvement in availability and
redundancy of power systems can significantly reduce the impact of less reduction in the average rate of major maintenance. In the ex-
unavailability of the power transformer. Fig. 7 provides a sensitivity ample, FRA does not reduce the operation cost because the cost savings
analysis under different cost settings. is offset by the annual operation cost of online monitoring. In practice,
Fig. 7 presents the operation cost when Cu,Cp,CM ′ and CR′ are scaled FRA is even less effective, because of the lack of interpretability and
down to 80%, 60%, 40%, 20% of the default setting. The inspection hence is usually applied only for post-fault tests.
interval is optimized to minimize the operation cost under each cost
setting. The minimum operation cost and optimal inspection interval
are presented in Table 2. 5. Conclusion
From Table 2, we can see that the minimum of operation cost de-
creases linearly when cost is reduced. In addition, the optimal inspec- In this paper, we first analyzed the malfunctions that may accelerate
tion interval increases non-linearly. the deterioration of winding insulation paper. Then, we applied the
We now examine how online condition monitoring might further information to refine the deterioration model of power transformers.
improve the performance of power transformer maintenance. We con- We considered multiple types of failure rates of power transformers,
sider two types of online monitoring devices: dissolved gas analysis such as infant mortality, sudden failure, and deterioration failures, so
device (DGA) and frequency response analysis device (FRA). that the overall failure rate of power transformer follows a bathtub
DGA determines the concentration of dissolved gas and moisture
content. It is very sensitive to the change of key gases produced by the Table 1
chemical aging of insulation paper. The accuracy of DGA can be as high Parameter setting [14,17,18,19].
as 90% [20]. It means that DGA may provide early warning for 90% of
Parameter Vale Parameter Value
malfunctions that cause accelerated chemical aging of the insulation
paper. Hence, the transition rate λ ol is: λa 0.04/year μM 24.39/year
λn 0.105/year μF 12.05/year
λ ol = 90% × λ f 1 = 0.0027/ year
λd1 5.333/year μR 3.04/year
λd2 2.105/year Cu £53,000/day
and λ f 1 is reduced to 0.0003/year. According [21] and [22], we assume
λf 1 0.003/year Cp £3200/day
the annually cost for operating a condition monitoring is:
λf 2 0.001/year ′
Cin £1000
Col = €3000/year ≈ £2550/year λF 0.008/year CC′ £1900
λFd 0.048/year CM′ £600,000
With the same method, we can calculate the availability and op- μin 1095/year CF′ £5600
eration cost by applying DGA. The results are illustrated in Figs. 8 and 9 μc 91.25/year CR′ £1,000,000
respectively.

179
Z. Liang, A. Parlikad Electrical Power and Energy Systems 99 (2018) 175–182

Fig. 5. Availability of the power transformer under different mean time between in-
spections.
Fig. 8. Availability of the power transformer under different mean time between in-
spections with DGA.

Fig. 6. Operation cost of the power transformer under different mean time between in-
spections.
Fig. 9. Cost of the power transformer under different mean time between inspections
with DGA.

′ and CR′ ).
Fig. 7. Sensitivity analysis under different cost setting (Cu,Cp,CM

Fig. 10. Availability of the power transformer under different mean time between in-
Table 2 spections with FRA.
Results from sensitivity analysis.

Percentage of the default cost Minimum operation Optimal inspection

curve with multiple wear-out tails. We developed analytical models for
setting cost interval
power transformer maintenance in the distribution network and
100% £89,750/year 0.3295 year transmission network using CTMC. A condition-based maintenance is
80% £72,414/year 0.353 year designed for power transformers in the transmission network. The de-
60% £55,017/year 0.392 year
signed condition-based maintenance model opens up a successful way
40% £37,503/year 0.467 year
20% £19,705/year 0.679 year to optimize the maintenance of power transformers according to the
information from both periodic inspection and online monitoring

180
Z. Liang, A. Parlikad Electrical Power and Energy Systems 99 (2018) 175–182

5
10 Table 3
1.25
Without FRA performance of DGA and FRA.
1.2 Optimal cost without FRA
With FRA
Optimal cost with FRA Online Benefit on Benefit on Reduction in the
Operation cost (£/year)

1.15 monitoring availability operation cost average rate of major

1.1 device maintenance

1.05 DGA 0.06% £3855/year 0.0015/year

FRA 0.01% −£2090/year 0.0002/year
1

0.95

0.9 Further work will be focused on extending this model to solve the
0.85 maintenance of a power system that is composed of multiples power
transformers, switches, and other devices.
0.8
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Mean time between inspections (year)
Acknowledgement
Fig. 11. Availability of the power transformer under different mean time between in-
spections with FRA.
We thank the reviewers for their constructive comments and prac-
tical insights that have helped us in improving the quality of this paper.
devices. The value of DGA and FRA analysis were studied in a numer- This research was funded by the Engineering and Physical Sciences
ical case. The result shows that by applying DGA, both the availability Research Council (UK) and Innovate UK through the Innovation and
and the operation cost of the power transformer can be improved. Knowledge Centre for Smart Infrastructure and Construction (Grant EP/
However, FRA can improve only the availability of the power trans- N021614/1).
former.

Appendix A

Expression of Steady States

State Expression State Expression

π0,0 1 π2,0 S2
π0,1 λf 0,1 π2,1 2 l λ di − 1,1 λf 2 − l,1
λ d0,1 + λFd
∑l = 0 ∏i = 1 S
λ di,1 + λFd λ d2 − l,1 + λFd 2 − l
π0,2 λf 0,2 π2,2 2 l λ di − 1,2 λf 2 − l,2
λ d0,2 + λFd
∑l = 0 ∏i = 1 S
λ di,2 + λFd λ d2 − l,2 + λFd 2 − l
π0,3 λF + λ a λFd 2 λf 0,j π2,3 λF λFd 2 2 l λ di − 1,2 λf 2 − l,2
μF
+ μF
∑ j=1 λ d0,j + λFd
S
μF 2
+ μF
∑ j = 1 ∑l = 0 ∏i = 1 S
λ di,2 + λFd λ d2 − l,2 + λFd 2 − l
π1,0 S1 π3,0 S3
π1,1 λ d0,1 λf 0,1 λf 1,1 π3,1 3 l λ di − 1,1 λf 2 − l,1
+ S
λ d1,1 + λFd 1
∑l = 0 ∏i = 1 S
λ di,1 + λFd λ d2 − l,1 + λFd 3 − l
∏1i = 0 λ di,1 + λFd
π1,2 λ d0,2 λf 0,2 λf 1,2 π3,2 3 l λ di − 1,2 λf 2 − l,2
+ S
λ d1,2 + λFd 1
∑l = 0 ∏i = 1 S
λ di,2 + λFd λ d2 − l,2 + λFd 3 − l
∏1i = 0 λ di,2 + λFd
π1,3 π3,3 λf 2 − l,2
( λF
μF
+ ∑ j=1
2 λf 1,j
λ d1,j + λFd )S + ∑
1
2 λ d0,j λf 0,j
j = 1 ∏1 λ + λ
i = 0 di,j Fd
λF
S
μF 3
+
λFd
μF
2 3 l
∑ j = 1 ∑l = 0 ∏i = 1
λ di − 1,2
S
λ di,2 + λFd λ d2 − l,2 + λFd 3 − l

where
2
λFd λ d0,j λf 0,j
λn0,0 + ∑ ∏1i = 0 λ di,j + λFd
j=1
S1 = 2
,
λFd λf 1,j
λn1,0 + ∑ λ f 1,j− λ
d1,j + λFd
j=1

2 2
λFd ∑
j=1
(∏ 2 λ di − 1,j
i = 1 λ di,j + λFd ) λf 0,j
λ d0,j + λFd
2
λni − 1,0 + ∑
j=1
λFd λ di − 1,j λfi − 1,j
(λ di,j + λFd )(λ di − 1,j + λFd )
S2 = 2
+ ∏i =1 2
λn2,0 + ∑
j=1
(λ f 2,j− λ
λf 2,j λFd
d2,j + λFd ) λni,0 + ∑
j=1
λ fi,j− λ
λFd λfi,j
di,j + λFd

and

181
Z. Liang, A. Parlikad Electrical Power and Energy Systems 99 (2018) 175–182

2 λf 0,j 2 λf 0,j 2 λFd λ d0,j λf 0,j

λ di − 1,j ⎞ λ di − 1,j
λFd ∑ ⎛⎜∏3i = 1 λ ⎟ λFd ∑ (∏3i = 1 λ ) λn0,0 + ∑ 1
di,j + λFd ⎠ λ d0,j + λFd di,j + λFd λ d0,j + λFd ∏ λ + λFd
j=1 ⎝ j=1 j = 1 i = 0 di,j
S3 = 2 + 2 2
λf 3,j λFd ⎞ λf 3,j λFd ⎞ λFd λf 1,j
λn3,0 + ∑ ⎛⎜λf 3,j − λ ⎟
⎛
λn3,0 + ∑ ⎜λf 3,j − ⎟ λn1,0 + ∑ λf 1,j −
d3,j + λFd λ d3,j + λFd λ d1,j + λFd
j=1 ⎝ ⎠ j=1 ⎝ ⎠ j=1
2 λFd λ di − 1,j λfi − 1,j 2 λf 0,j 2 λFd λ d2,j λf 2,j
λ di − 1,j ⎞
λni − 1,0 + ∑ λFd ∑ ⎜⎛∏i2= 1 λ ⎟ λn2,0 + ∑
(λ di,j + λFd)(λ di − 1,j + λFd) di,j + λFd ⎠ λ d0,j + λFd ∏3i = 2 λ di,j + λFd
3 j=1 j=1 ⎝ j=1
+ ∏i = 1 2 + 2 2
λFd λfi,j λf 2,j λFd λFd λf 3,j
λni,0 + ∑ λfi,j − λ λn2,0 + ∑ (λf 2,j − λ ) λn1,0 + ∑ λf 3,j − λ
di,j + λFd d2,j + λFd d3,j + λFd
j=1 j=1 j=1

References
[1] Bartley WH. Analysis of transformer failures. In: International association of en-
gineering insurers 36th Annual Conference; 2003, p. 1–5.
[2] Allan DM. Insulation ageing-a users perspective. In: Properties and applications of
dielectric materials, 1994, Proceedings of the 4th International Conference on,
1994, vol. 2, p. 833–6.
[3] Aldhubaib HA, Salama MM. A novel approach to investigate the effect of main-
tenance on the replacement time for transformers. IEEE Trans Power Deliv
2014;29(4):1603–12.
[4] Dhople SV, Chen YC, Domínguez-García AD. A set-theoretic method for parametric
uncertainty analysis in Markov reliability and reward models. IEEE Trans Reliab
2013;62(3):658–69.
[5] Abu-Elanien AE, Salama MMA, Bartnikas R. A techno-economic method for repla-
cing transformers. IEEE Trans Power Deliv 2011;26(2):817–29.
[6] Lima SL, Saavedra OR, Miranda V. A two-level framework to fault diagnosis and
decision making for power transformers. IEEE Trans Power Deliv
2015;30(1):497–504.
[7] Abiri-Jahromi A, Parvania M, Bouffard F, Fotuhi-Firuzabad M. A two-stage frame-
work for power transformer asset maintenance management—Part I: models and
formulations. IEEE Trans Power Syst 2013;28(2):1395–403.
[8] Koksal A, Ozdemir A. Improved transformer maintenance plan for reliability
centred asset management of power transmission system. IET Gener Transm Distrib
2016;10(8):1976–83.
[9] Hohlein I, Kachler AJ. Aging of cellulose at transformer service temperatures. Part
2. Influence of moisture and temperature on degree of polymerization and forma-
tion of furanic compounds in free-breathing systems. IEEE Electr Insul Mag
2005;21(5):20–4.
[10] Liao R, Xiang B, Yang L, Tang C, Sun H. Study on the thermal aging characteristics
and bond breaking process of oil-paper insulation in power transformer. In:
Electrical Insulation, 2008. ISEI 2008. Conference Record of the 2008 IEEE inter-
national symposium on, 2008, p. 291–6.
[11] Prevost TA, Oommen TV. Cellulose insulation in oil-filled power transformers: part
I-history and development. IEEE Electr Insul Mag 2006;22(1):28–35.
[12] Lundgaard LE, Hansen W, Ingebrigtsen S. Ageing of mineral oil impregnated cel-
lulose by acid catalysis. IEEE Trans Dielectr Electr Insul 2008;15(2):540–6.
[13] Poan X, Zhishu Z, Macalpine M, Zhijian J, Zhushi R. Estimation of the clamping
pressures of transformer windings from vibration measurements. HKIE Trans
2008;15(1):30–5.
[14] Sokolov V, Bulgakova V, Berler Z. Assessment of power transformer insulation
condition. In: Electrical insulation conference and electrical manufacturing coil
winding conference, 2001. Proceedings, 2001, p. 605–13.
[15] Emsley AM, Stevens GC. Review of chemical indicators of degradation of cellulosic
electrical paper insulation in oil-filled transformers. IEE Proc.-Sci. Meas. Technol.
1994; 141(5), p. 324–34.
[16] Emsley AM, Heywood RJ, Ali M, Xiao X. Degradation of cellulosic insulation in
power transformers. Part 4: effects of ageing on the tensile strength of paper. IEE
Proc.-Sci. Meas. Technol. 2000; 147(6), p. 285–90.
[17] Mijailovic V. Method for effects evaluation of some forms of power transformers
preventive maintenance. Electr Power Syst Res 2008;78(5):765–76.
[18] Westman T, Lorin P, Ammann PA. Fit at 50 – keeping aging transformers healthy for
longer. Asset Manag Maint J Apr. 2011;24(2):38.
[19] Liang Z, Parlikad AK. A condition-based maintenance model for assets with ac-
celerated deterioration due to fault propagation. Reliab IEEE Trans On
2015;64(3):972–82.
[20] Muhamad NA, Phung BT, Blackburn TR, Lai KX. Comparative study and analysis of
DGA methods for transformer mineral oil. In: Power Tech, 2007 IEEE Lausanne;
2007, p. 45–50.
[21] Boss P, Lorin P, Viscardi A, Harley JW, Isecke J. Economical aspects and practical
experiences of power transformers on-line monitoring. Int. Counc. Large Electr.
Syst. CIGRÉ Fr., 2000.
[22] Boss P, Horst T, Lorin P, Pfammatter K, Fazlagic A, Perkins M. Life assessment of
power transformers to prepare a rehabilitation based on a technical-economical
analysis. In: Cigre Session; 2002, p. 12–106.
[23] Kim J-W, Park B, Jeong SC, Kim SW, Park P. Fault diagnosis of a power transformer
using an improved frequency-response analysis. IEEE Trans Power Deliv
2005;20(1):169–78.

182